MJ Statistical Regulator PCBs

[Cressy Snr]

I'll be one of your testers Nick.
I can do them as screen reg(s)
I can build one of these, even one per channel (which might be a good idea) for the input stage screens of my amp.
I'm already half way there with a 100V string of diodes without a ccs , so it wouldn't be difficult to build your board
and test and evaluate it in my amp as I already have a good idea of what's involved, having got a partial one in there already.

[simon]

Like Paul I'm a long way from being able to test unfortunately, but a big thumbs up for doing this

[cressy]

i could build and test too jack, i have either the phono stage or my main amp i can slot one into, the phono stage pulls about 20 ma so would probably be a better test subject, i cant remember what the front end of the amp pulls tbh

[Ray P]

I have a new digital source to build that will need 12mA @ 240V, which is why I've expressed interest, however, I know a key part of it won't be available for at least a couple of months so will be unable to evaluate the results. I'll have to leave the build/test to others. Sorry.

Ray

[thomas]

Hi
Yes I can test one of these for you Nick, can use it to power a phonostage, 250v ish, approx 12mA IIRC….

[jack]

OK - thanks guys - three is probably plenty at this stage.

I'll send you each a PM...

Ta

[jack]

Kits went out today - Steve's & Ant's are in the same package...

[thomas]

Hi, got mine today, ta muchly….

I may be speaking too soon but those comprehensive building instructions you've written look suitably idiot proof…..

Buuuut….I'm slightly confused why your circuit diagram has 'Input<420v' at the top, but then suggests using an IXTP08N100D2 for input voltages above 340v instead of the DN2540N5 - i'll have to look that device up, actually….


Anyhow, will put it together this weekend….

Well done again for putting it together

T

[Cressy Snr]

Yep, got mine too.
Will put it together this W/E and thanks for turning the MJ statistical reg into an easy to build (I think) practical format.

[jack]

I may be speaking too soon but those comprehensive building instructions you've written look suitably idiot proof…..

Buuuut….I'm slightly confused why your circuit diagram has 'Input<420v' at the top, but then suggests using an IXTP08N100D2 for input voltages above 340v instead of the DN2540N5 - i'll have to look that device up, actually….

The notes on the schematic are really a stream-of-consciousness on my part - as it happens, the DN2540 should be fine for pretty much any use case as its heavily over-specced - T1 will only see the difference between Vin and Vout, typically < 100V (hopefully < 50V), so thanks for pointing that out - the comment originally was before I worked out how to easily use the expansion board to take the range way way up....

The only time you might get a big PD across the CCS is when using the board as a pure CCS as mentioned in the documentation - its theoretically possible but unlikely.

The IXT device is another possible candidate for the future as a replacement for the DN2540, but not tried yet.

[Cressy Snr]

OK,
Have got my module built.
The instructions are excellent; clear and user friendly.

The folding over of the Zener diode cathode leads explanation is great and ensures that unless you are a complete moron, you can't go wrong.

I cut the cathode leads of all the diodes I wanted whilst they were still on the tape strips leaving the anode ends attached to their strip. Then taking a six inch plastic ruler as a jig I folded the cathode leads over the ruler edge so that they were all identical. You can then trim off the anode leads next to their tape strip, leaving all the zeners hanging over the ruler, diode end stripe uppermost.
Lean the ruler on something solid and you can simply take each zener off one at at time and insert it into the board, soldering the cathode leads in from the top as you go. That way the correct orientation is taken care of for you and all you need to do is make sure that the diode body, sits on its white circle. Circular diode body sits on circular mark....easy.
Once they are all in, the board can be turned over, leads trimmed and the round (anode) ends all soldered up on the underside.

Only thing I did wrong was to solder one of the test jack pin sets in upside down, which once you do it you're stuffed. Sod's law meant I didn't notice until I had soldered both bloody pins, otherwise I could have retrieved the situation. So now I'm going to have to solder a link across the now too short pins once testing and set-up is done. Not a serious mistake but something to watch out for.
Perhaps in the instructions Nick, you could mention the need for the builder to make damned sure that the LONG (gold) ends of the test pin assembly are sticking up above the board before said assembly is soldered in place. A silly mistake, but I managed to make it, so a line about watching out for that error might be something to tag onto point 12 in the instructions.

I haven't tested the board yet but I'll report, when I've done that, later this weekend.

[jack]

Thanks for that, Steve.

Regarding the inverted pins - you could always use some side-cutters to snip the connecting plastic bit and then get the pins out individually....

I'll update the docs as you suggest

Cheers

[Cressy Snr]

OK..
The module worked first time and with the zener string set to 12mA, it regulated the screens of the input stage 12SJ7s at a rock solid 103V; not even the slightest variation. I set the regulator input voltage to 127V for 103V output.
The voltage drop from the main 392V HT, down to the 127V statistical reg input, was set by means of a pair of VR150 and VR105 glow tubes in series. It was handy that I hadn't removed the octal sockets.

Soundwise, my ears pricked up with the first track I played (Come Tomorrow by Manfred Mann). There was a studio with width and depth behind Paul Jones and the voice was spot on both emotionally and tonally. I'll listen to a few more pieces of music tomorrow and give a more detailed report, but on the evidence so far, as a screen regulator, the MJStR looks as if it is going to be a winner.

EDIT: Just a small detail point again. As the final adjustments to the reg have to be carried out on the working circuit, It would be good to have the blue adjustment pot for the CCS with the adjusting screw on the top rather than on the side. In my valve amp, the thought of adjusting the current, with caps charged to 400V at 150 Joules, in the vicinity of my right hand gave me the willies

I ended up placing a hard plastic piece of sheet material over the cap bank terminals before I dared go anywhere near the adjuster screw. Sometimes a little fear can be a good thing

A top adjuster would have made life that little bit easier and a helluva lot safer.

[jack]

EDIT: Just a small detail point again. As the final adjustments to the reg have to be carried out on the working circuit, It would be good to have the blue adjustment pot for the CCS with the adjusting screw on the top rather than on the side. In my valve amp, the thought of adjusting the current, with caps charged to 400V at 150 Joules, in the vicinity of my right hand gave me the willies

Extremely good point Steve - I should have thought of that - just like for the Bournes ranges, Vishay have a vertical adjust version of the M64X101KB40 called the M64W101KB40, which has the same profile, footprint and cost.

I'll offer purchasers the choice of top (vertical) or side (horizontal) adjustment...

I'll listen to a few more pieces of music tomorrow and give a more detailed report, but on the evidence so far, as a screen regulator, the MJSR looks as if it is going to be a winner.

BTW. I presume you are using 18 zeners?

[Cressy Snr]

BTW. I presume you are using 18 zeners?

Yep, 18 zeners.